Soft Bit Metric Generation Using Dominant LLR Terms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Differential encoding techniques, such as DE-QPSK, face high computational complexity in generating soft bit metrics due to the numerous terms required in reliability calculations, which are essential for accurate data recovery in noisy channels.
Innovation Solution
Selecting dominant terms from the expression of reliability for encoded bits in higher order modulation schemes, specifically using the G operation and correction terms to reduce computational complexity, and implementing this in circuitry or software for efficient soft bit metric generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If soft bit metric information is generated using full reliability expressions with all terms, then decoding accuracy is improved, but computational complexity increases significantly
Solution Approach 1:
The patent extracts only the dominant terms from the full reliability expression for soft bit metric generation. By identifying and removing non-dominant terms that contribute minimally to the final result, the system maintains decoding accuracy while significantly reducing the number of calculations required. This is achieved by analyzing the reliability expression and selectively retaining only those terms that have substantial impact on the soft metric values.
Solution Approach 2:
The patent changes the parameter selection in the reliability expression by introducing a threshold or significance criterion. Instead of using all terms with equal weight, the system modifies which terms are included based on their dominance or significance level. This parameter change allows the system to adaptively select terms, reducing computational load while preserving the essential information needed for accurate decoding.
2Measurement precision
If all terms in the reliability expression are calculated, then reliability estimation accuracy is improved, but processing time increases
Solution Approach 1:
The patent extracts and calculates only the dominant terms from the reliability expression, eliminating the need to compute all terms. By identifying which terms dominate the reliability calculation and focusing computational resources on those specific terms, the system achieves accurate reliability estimation with significantly reduced processing time. Non-dominant terms are excluded from calculation entirely.
Solution Approach 2:
The patent applies partial action by calculating only the necessary subset of terms required for accurate reliability estimation, rather than computing the complete set. This selective approach performs just enough calculation to achieve the required accuracy level, avoiding the excessive processing time that would result from calculating all terms in the full expression.
3Productivity
If dominant terms are selected and used for reliability calculation, then computational efficiency is improved, but potential accuracy loss may occur
Solution Approach 1:
The patent carefully extracts only the truly dominant terms that have significant impact on reliability accuracy. By using rigorous criteria to identify which terms are genuinely dominant versus which are merely present, the system ensures that removing non-dominant terms does not compromise accuracy. The extraction process is designed to retain all terms that contribute meaningfully to the final reliability metric.
Solution Approach 2:
The patent adjusts the selection parameters and thresholds for term inclusion to optimize the balance between computational efficiency and accuracy. By carefully tuning which terms are classified as dominant and should be retained, the system maximizes computational efficiency while maintaining reliability accuracy. The parameter settings ensure that only terms with minimal impact are excluded.
Data Source
AI summary
Soft bit metric generation computational complexity can be reduced by identifying and utilizing only the dominant terms in a reliability calculation such as a logarithmic likelihood ratio (LLR). The dominant terms are those terms for which the signs of the x and y components match those of channel outputs of the channel outputs. One technique for identifying the dominant terms is by determining the most likely transitions from two consecutive channel output samples Values for the dominant terms can be estimated by either the joint reliability of two consecutive samples of the in-phase component (x1,x2) or by the joint reliability of two consecutive samples of the quadrature components (y1,y2).